optmatch: Functions for Optimal Matching

Provides routines for distance based bipartite matching to reduce covariate imbalance between treatment and control groups in observational studies. Routines are provided to generate distances from GLM models (propensity score matching) and formulas (Euclidean and Mahalanobis matching), stratified matching (exact matching), and calipers. Results of the fullmatch routine are guaranteed to provide minimum average within matched set distance.

Version: 0.9-5
Depends: R (≥ 2.15.1), stats, methods, graphics, survival
Imports: Rcpp, RItools, digest
LinkingTo: Rcpp
Suggests: boot, biglm, testthat, brglm, arm
Published: 2015-07-31
Author: Ben B. Hansen, Mark Fredrickson, Josh Buckner, Josh Errickson, and Peter Solenberger, with embedded Fortran code due to Dimitri P. Bertsekas and Paul Tseng
Maintainer: Mark M. Fredrickson <mark.m.fredrickson at gmail.com>
License: file LICENSE
URL: http://www.r-project.org, https://github.com/markmfredrickson/optmatch
NeedsCompilation: yes
Citation: optmatch citation info
Materials: NEWS
In views: Optimization, SocialSciences
CRAN checks: optmatch results


Reference manual: optmatch.pdf
Package source: optmatch_0.9-5.tar.gz
Windows binaries: r-devel: optmatch_0.9-5.zip, r-release: optmatch_0.9-5.zip, r-oldrel: optmatch_0.9-5.zip
OS X Snow Leopard binaries: r-release: optmatch_0.9-5.tgz, r-oldrel: optmatch_0.9-3.tgz
OS X Mavericks binaries: r-release: optmatch_0.9-5.tgz
Old sources: optmatch archive

Reverse dependencies:

Reverse suggests: MatchIt, rcbalance, RcmdrPlugin.EZR
Reverse enhances: RItools